So I thought I'd kick off some discussion about table flatness, consistency and versatility. I've learned a few things over the past month that I didn't know/encounter much before and thought I'd share some of them. Feel free to jump in and compare notes, ask questions etc. My company, among other things, is classified as a job shop. I machine just about any material in 2D and 3D, so versatility is important. By versatility I mean that sometimes I can get away with holding things down using some kind of clamps, screws or adhesive, while other times I need vacuum - and even less, but not out of the realm of possibility I have to bolt things down.

My shop is between two rivers and weather-wise I encounter everything from hot & humid to dry and sub-freezing cold. As you probably know, every material you put your hands on has a coefficient of expansion, and wood-based materials also change size by humidity. Material also moves around while it is being machined because it is being heated up by the action of cutting and the surrounding shop air which is also heating up. This is why I never stop a long toolpath and go to sleep because it WILL change dimension overnight as things cool down. What I didn't know was how much an ultralight MDF bleeder board could change in size during a long cycle time (13hrs).

My machine, 'Big Bertha', is a 5x16' machinable area PRT Alpha (7.2:1 all around). It started out as a custom tool with dual Zs, 5hp spindle on Z1 w/5x8' spoilboard and a Hypertherm plasma cutter on Z2 with 5x8' steel grid. I've since done away with the plasma and my current machinable area is 5x12 - if for no other reason than I can't get sheet goods larger than this. I've got a 2nd gantry parked down at the 16' end which I am going to bring back online. My current setup consists of 5x12' 3/4" MDO plywood as the support board, regular MDF for the vacuum grid (zoned out accordingly) and 5x12x 3/4" Trupan Ultralight MDF for the bleeder board. All layers are glued down including the bleeder (on top of grid squares AND zone perimeters)

In days gone by, I used my own vacuum plenum design (BradyVac removable sheets) made out of Trupan UL and for a while a Fein Turbo III (@7Hg") as the vac source. When used properly, it worked pretty well, and still (via screws) gave me access to the MDF table underneath for screws, tape etc. It was OK when the vacuum was on in terms of being relatively flat and I could machine it flat with the tool. When I moved things to the new shop, I did the dual vac setup, which allows me to select between a 5hp FPz (9.5 Hg") and 10hp Becker (25.5 Hg") depending on what I am doing. Hey, why run 10hp when you can run 5?

Ever since doing a fully dedicated vacuum system (which I really wrestled with - in terms of giving up some versatility) - I've had issues with the table not being as flat as I was accustomed to. My previous layup was MDO & Medex water resistant MDF - which I think they changed the formula on...and it was ROCK SOLID in terms of not moving around and staying flat. Now I'm finding myself flattening the bleeder more often than I think I should just to get things to be flat...for a little while (as in only hours). I recently did a 8'+ long filagree panel in 3D which took 13hrs to complete on the finishing pass alone. After doing a 3D roughing pass that left .03" allowance on it, I finish machined it with a 3/16" ball @ 8% because it had a lot of compound curves to reduce tool marks. About 1/2 way through the finishing pass, I was seeing that it was barely taking off that .03" of allowance - which meant that the entire table was sinking in...I found myself doing some black magic, stopping the tool every .05", nudging the Z down .003" each time until I was able to get that .03" back. When all was said and done, the table had sucked down a full .14" (YIKES!!!) across only 24". There's no way I could have put my name on that piece so it's hanging up on the wall of shame...

I wound up machining off what was left of the bleeder board and machining the grid flat. I glued down another piece of Trupan Ultralight and then machined it flat. I did the entire job over only to find to my extreme displeasure that it was doing the EXACT SAME THING - albeit a little less at .05" creep. I was able to fudge things a bit and it all worked out - but the 'builder's curse' still remains where I know all that is wrong with it even if the customer doesn't. (They literally high-fived each other in the office - which is next door to the Ferarri dealership...when they saw it - which erased some of my shame.)

So...I made some calls to some good friends who also do CNC routing professionally and we talked about the issues. My one buddy's shop is in Miami and he routinely cut 40-60 sheets at a time. He told me that it wasn't uncommon for him to replace his bleeder every 3 days (!!!) because the bleeder would either swell up, compress under the vacuum suction or dry out and change dimension in the winter - or a combination thereof. He was running a few Lighthouse vacs (9Hg")and was having this happen...it could only be worse with higher Hg". So what do the $500k+ CNC router machining centers use then? How do they get around this problem? Density...and more vacuum.

The big machines use several Becker vacuums (or similar) that pull 25 Hg"+ to get the CFM up AND they use regular old MDF. When they have all vac zones open, they pull 12 Hg" or so with nothing on top of the table - for a delta/useable vacuum range of about 13 Hg" out of their max 25 Hg". 13 Hg" is usually plenty to hold down sheet goods through a bleeder. THis makes sense that the higher density MDF will be more stable - and if I really take a look at the Trupan UL, it is about as free-flowing as anything out there. It would probably make a good air filter! So I think I'm all done with ULMDF - which is OK for guys running the Lighthouse/Fein setups. It just crushes down so easily under the suction of the Becker. It also must absorb a lot of moisture and in turn shrink from getting sucked out.

Just because I've been doing this for a while doesn't mean I don't get schooled from time to time The important part is learning and moving forward. One interesting thing is the little SB Desktop machine. I've had that for (I think) about 4 years now. It has the Bosch extrusion - same stuff used on the PRS table sides, as the 'support board' with a piece of 3/4" MDF bolted down every foot or so, with countersunk bolts. Now it's just a small machinable area (18x24") but that spoilboard is totally rock solid. When I flatten it, I only have to take off like .01-.015" and it's dead flat again with all kerf traces gone. It's simply amazing compared to what I go through with the big tool.

So I am strongly considering doing a similar setup on the 5x16. The MDO support board goes away & gets replaced with some aluminum bar that adapts the steel crossmembers to 1030 8020 aluminum extrusions - all the way across the 60" bed (20pcs). This alone will function as a torsion box, and I can get it as dead flat as possible using a dial indicator on the gantry. Parts to be machined can be clamped or bolted down using any number of clamps/fixures including gasketed high-Hg" vacuum pods.

On top of the extrusions, a regular MDF table, consisting of 1-man manageable sheets/tiles can be bolted down just like the little DT machine. Parts can be screwed down with a nice soft MDF board to protect the cutter. There's no more need to mess with big 5x12' sheets - they can be all put together and made into one big sheet as needed and replaced individually.

To keep the versatility of a universal vacuum system, the same system can be used. The MDF tiles can be removed and replaced with a BradyVac (BradyVac3 anyone?) I plan to use regular 3/4" MDF as the bleeder - machined BradyVac style about 1/4" deep for the grid. Then I'll glue on a 1/4" MDF backer with 2" hole for the vacuum port and flip the whole thing over. The same countersunk bolt scheme will be used (bolt holes coincident with 1" squares every 12" or so) along with o-rings under the fastener heads. The MDF backer can be sealed where it meets the extrusion and edges. Of course there will need to be a 3x3" piece of extrusion missing where the vacuum ports are...but no big deal.

See attached pics below...tell me what you think...what you'd change/add/do differently. It's not a cheap endeavor @ around $2k for the extrusions, but confidence in the machine is paramount and I don't think I'd rest easy until I've done my best to get positive control over the situation.

-B

1x3ExtrudedBase.jpgCountersunkMDF_BoltedToExtrusion.jpgInvertedBradyVacGridwithQuarterInBackerBoltsDown2.jpg